Loading...

​​Journocoders January 2019

​​An introduction to analysing data and text with R

​​

​​Show and tell

​​Read anything related to data or journalism recently? Please add links here! Stories, analysis, announcements -- anything you think others might be interested in.

• ​​https://mapschool.io/

• ​​https://gijn.org/2018/11/20/how-satellite-imagery-became-an-indispensable-and-easily-accessible-tool-for-journalists/?mc_cid=b2a0216003&mc_eid=672c9843e1

• ​​https://gijn.org/2018/12/27/the-25-best-data-visualizations-of-2018/

• ​​https://neo4j.com/graphacademy/online-training/introduction-to-neo4j/

• ​​https://www.dropbox.com/s/h8ejagx2e8z3fr8/IMG_0871.PNG?dl=0

• ​​http://www.maartenlambrechts.com/2018/12/31/the-list-of-2018-visualization-lists.html

• ​​
  

​​

​​Tutorial

​​You will need a couple of packages installed before starting the tutorial. To do this inside R Studio enter:

​​

​​install.packages(c('tidyverse', 'tidytext', 'maps'))

​​

​​It might take a few minutes to complete. You will also need to download the data from here. Then you need to tell R Studio where to look for the data. In the folders list on the bottom right of the R Studio window find the winemag-data.csv​ file you just downloaded, then click the More​ button just above, and then Set As Working Directory​.

​​

​​Then follow the main tutorial: http://www.storybench.org/using-french-wine-reviews-understand-tf-idf-measure-unique-word-document/

​​

​​You can fix the incorrect character in 'Rhône Valley' by adding encoding = "UTF-8"​ to the read.csv​ parameters:

​​dataset <- read.csv("winemag-data.csv",

​​                    header = TRUE,

​​                    stringsAsFactors = FALSE,

​​                    encoding = "UTF-8")

​​

​​If you are having a problem with this bit

​​

​​counts <- francewines %>% 

​​group_by(province) %>% 

​​summarise(lat = lat, lon = lon, count = n(),

​​            mean_price = mean(price, na.rm = TRUE),

​​            mean_points = mean(points)) 

​​counts %>% glimpse()

​​

​​Replace it with this

​​

​​counts <- francewines %>% 

​​  group_by(province) %>% 

​​  summarise(lat = mean(lat), lon = mean(lon), count = n(),

​​            mean_price = mean(price, na.rm = TRUE),

​​            mean_points = mean(points)) 

​​

​​Next time

​​The next Journocoders meetup will be on Thursday 14th February at the News Building.

​​

​​What should we cover? Anything we should or shouldn’t be doing? We want your ideas!

​​